SANDIA REPORT SAND2014-15239 Unlimited Release Printed Month and Year Offshore Wind Guidance Document: Oceanography and Sediment Stability (Version 1) Development of a Conceptual Site Model Jason Magalen, Craig Jones, and Jesse Roberts Prepared by Sandia National Laboratories Albuquerque, New Mexico 87185 and Livermore, California 94550 Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Approved for public release; further dissemination unlimited.
122
Embed
Offshore Wind Guidance Document: Oceanography and Sediment ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
SANDIA REPORT SAND2014-15239 Unlimited Release Printed Month and Year
Offshore Wind Guidance Document: Oceanography and Sediment Stability
(Version 1)
Development of a Conceptual Site Model
Jason Magalen, Craig Jones, and Jesse Roberts
Prepared by Sandia National Laboratories Albuquerque, New Mexico 87185 and Livermore, California 94550
Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Approved for public release; further dissemination unlimited.
2
Issued by Sandia National Laboratories, operated for the United States Department of Energy
by Sandia Corporation.
NOTICE: This report was prepared as an account of work sponsored by an agency of the
United States Government. Neither the United States Government, nor any agency thereof,
nor any of their employees, nor any of their contractors, subcontractors, or their employees,
make any warranty, express or implied, or assume any legal liability or responsibility for the
accuracy, completeness, or usefulness of any information, apparatus, product, or process
disclosed, or represent that its use would not infringe privately owned rights. Reference herein
to any specific commercial product, process, or service by trade name, trademark,
manufacturer, or otherwise, does not necessarily constitute or imply its endorsement,
recommendation, or favoring by the United States Government, any agency thereof, or any of
their contractors or subcontractors. The views and opinions expressed herein do not
necessarily state or reflect those of the United States Government, any agency thereof, or any
of their contractors.
Printed in the United States of America. This report has been reproduced directly from the best
Offshore Wind Guidance Document: Oceanography and Sediment Stability
Jason Magalen and Craig Jones
Sea Engineering, Inc.
200 Washington Street, Suite 101
Santa Cruz, CA 95060
Jesse Roberts
Water Power Technologies
Sandia National Laboratories
P.O. Box 5800
Albuquerque, New Mexico 87185-MS1124
Abstract
This guidance document provides the reader with an overview of the key environmental
considerations for a typical offshore wind coastal location and the tools to help guide the reader
through a thorough planning process. It will enable readers to identify the key coastal processes
relevant to their offshore wind site and perform pertinent analysis to guide siting and layout
design, with the goal of minimizing costs associated with planning, permitting, and long-term
maintenance. The document highlights site characterization and assessment techniques for
evaluating spatial patterns of sediment dynamics in the vicinity of a wind farm under typical,
extreme, and storm conditions. Finally, the document describes the assimilation of all of this
information into the conceptual site model (CSM) to aid the decision-making processes.
4
ACKNOWLEDGMENTS
The research and development described in this document was funded by the U.S. Department of
Energy. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin
Company, for the United States Department of Energy’s National Nuclear Security
Administration under contract DE-AC04-94AL85000.
This research was made possible by support from the Department of Energy’s Wind and Water
Power Technologies Office.
5
CONTENTS
1. Introduction .............................................................................................................................. 11 1.1. Background ................................................................................................................... 11 1.2. Current Status in the United States ............................................................................... 12 1.3. Using This Document ................................................................................................... 13
2. Guidance and recommendations .............................................................................................. 17 2.1. Lessons Learned and Existing Guidance ...................................................................... 17 2.2. EIA General Guidance (CEFAS, 2004) ........................................................................ 19 2.3. Oceanographic and Sediment Dynamic Specific Considerations ................................. 21 2.4. Modeling Considerations (Physical and/or Numerical) ................................................ 22
2.5.1. MSP in the U.S. .............................................................................................. 23
3. Sediment Transport Fundamentals .......................................................................................... 25 3.1. Introduction ................................................................................................................... 25
3.2. Role of Hydrodynamic Processes ................................................................................. 27 3.3. Physical Properties of Sediment ................................................................................... 29
3.4. Sediment Erosion .......................................................................................................... 30 3.5. Sediment Transport ....................................................................................................... 32
3.5.1. Non-Cohesive Sediment Transport .................................................................... 33
4.5. Scour Potential – Cables and Pipes .................................................................................. 53 4.5.1. Sand Burial ......................................................................................................... 53 4.5.2. Water Depth ....................................................................................................... 54
4.5.3. Scour in Currents ................................................................................................ 54 4.5.4 Scour in Waves .................................................................................................... 55 4.5.5. Scour in Waves and Currents ............................................................................. 56
6
4.5.6. Time Development of Scour .............................................................................. 56
Distribution ................................................................................................................................. 122
FIGURES
Figure 1. Offshore wind specific CSM flowchart example. ........................................................ 15 Figure 2. Examples of offshore wind turbine foundation types (www.offshorewind.net). ......... 26 Figure 3. Udden-Wentworth grain size scale for sediments. ....................................................... 28 Figure 4. Simplified diagram of sediment transport processes. ................................................... 30 Figure 5. Simplified diagram of sediment transport processes. ................................................... 31 Figure 6. Cohesive aggregates eroded from the bed may disaggregate downstream. ................. 32
7
Figure 7. Shields curve for the initiation of motion for steady flow (from ASCE 1975). ........... 34
Figure 8. Threshold of motion of sediments beneath waves and/or currents. The fitted curves of
both Shields (1936) and Soulsby and Whitehouse (1997) are shown. ......................................... 35 Figure 9. Illustration of longshore current generation and resultant sediment transport. ............ 39
Figure 10. Diagram of the primary, cross-shore, wave driven current in the near shore. ........... 40 Figure 11. Illustration of sediment transport processes interacting in the offshore coastal zone. 42 Figure 12. Hydrodynamics around a slender pile with scour. ..................................................... 44 Figure 13. Example of local (clearwater) and global scour around a structure ............................ 45 Figure 14. Tunnel scour causing erosion of sediments beneath pipe/cable (Sumer and Fredsoe,
2002). ............................................................................................................................................ 47 Figure 15. Free-span showing the span shoulders supporting the pipe/cable. .............................. 47 Figure 16. Lee-wake vortex shedding in lee of a pipe/cable span (Sumer and Fredsoe, 2002). .. 47 Figure 17. Potential scour modes occurring in the vicinity of a transmission line. Scour can
propagate longitudinally along the line causing large unsupported spans. ................................... 48 Figure 18. Illustration of the coastal zone as a component of the Conceptual Site Model. .......... 67
Figure 19. Series of littoral cells along the California coast ......................................................... 68 Figure 20. Elements to consider in the near-shore sediment budget. ........................................... 69
Figure 21. Map of shoreline stability assessment (NASA)........................................................... 71 Figure 22. Illustration of single- (left) and multi-beam swath (right) survey technology bottom
Figure 24. Example meteorological and oceanographic data from the mouth of Chesapeake Bay
(NOAA). ....................................................................................................................................... 74 Figure 25. Example spectral wave data from a buoy coupled with a real time coastal wave model
Figure 26. Solid support systems used for offshore wind applications: (a) monopile structure; (b)
suction caisson multi-foundation structure; (c) a suction caisson monopod structure. ................ 82 Figure 27. Example gravity based foundation (http://www.eon-uk.com). ................................... 83
Figure 28. Top panel shows a general catenary moored platform where the tethers are arced.
Bottom panel shows an exaggerated general tension leg moored platform where the tethers are
drawn taut...................................................................................................................................... 84 Figure 29. Pictorial representation of a moored floating wind turbine. The main components of
the support system are the nacelle, platform, ballast and mooring system. .................................. 84 Figure 30. General CSM flowchart. .............................................................................................. 98 Figure 31. Offshore wind specific CSM flowchart example. ....................................................... 99 Figure 32.Monterey Bay model domain and bathymetry. NOAA NDBC buoys used for model
validation are shown in green. .................................................................................................... 111
Figure 33.Model (line) representing the wave height (Hs), peak wave period (Tp) and mean wave
direction (MWD) obtained from the Monterey Bay SWAN model. Measured data (dots) were
obtained from the NOAA NDBC buoy 46236 in Monterey Bay. .............................................. 112 Figure 34. Model (line) representing the current magnitude obtained from the nearshore Santa
Cruz EFDC model. Measured data (dots) were obtained from a Teledyne RDI ADCP deployed
during the field study. ................................................................................................................. 112 Figure 35. Peak wave heights modeled using SWAN in the Santa Cruz, CA region. ................ 113
8
Figure 36.Peak wave heights and velocity vectors in the model domain. Expanded view of the
above region. ............................................................................................................................... 114 Figure 37. Monterey Bay model domain. 200 offshore wind turbine array. .............................. 115 Figure 38. Modeled wave heights before (top) and after (bottom) the inclusion of a wind turbine
array. ........................................................................................................................................... 116 Figure 39. Modeled seabed shear stresses before (top) and after (bottom) the inclusion of a wind
turbine array. ............................................................................................................................... 119 Figure 40. Risk of sediment transport before (top) and after (bottom) the installation of a wind
Figure 41.Flowchart of risk assessment methodology. ............................................................... 121
TABLES
Table 1. Shape factor, Ks, compiled by Melville and Sutherland (1988). .................................... 51 Table 2. Data needs to develop site characterization. ................................................................... 65 Table 3. Near-field processes and effects. .................................................................................... 90
Table 4. Far-field processes and effects. ....................................................................................... 91 Table 5. Matrix to evaluate the relative importance of an impact. ............................................... 93
9
NOMENCLATURE
ABS Acoustic Backscatter
CDIP Coastal Data and Information Program
CEFAS Center for Environment, Fisheries, and Aquacultural Science
COWRIE Collaborative Offshore Wind Research Into the Environment
CSM Conceptual Site Model
DEM Digital Elevation Model
DNV Det Norske Veritas
DOE Department of Energy
DOI Department of Interior
EIA Environmental Impact Assessment
ETSU Energy Technology Support Unit
EWEA European Wind Energy Association
GBF Gravity Based Foundation
GIS Geological Information System
GPS Global Positioning System
GW Gigawatts
KC Keulegan Carpenter
MSP Marine Spatial Planning
MW Megawatts
NDBC National Data Buoy Center
NGDC National Geodetic Data Center
NOAA National Ocean and Atmospheric Administration
NWS National Weather Service
OSWinD Offshore Wind Innovation and Demonstration
SEDFlume Sediment Erosion at Depth Flume
SNL Sandia National Laboratories
SSC Suspended Sediment Concentration
UK United Kingdom
US United States
USACE US Army Corp of Engineers
USGS US Geological Survey
VIV Vortex Induced Vibrations
10
11
1. INTRODUCTION
1.1. Background
Development of alternative energy production methods in the United States continues at a rapid
pace, with significant public and private investment in recent years. Though some are proven
energy-generating technologies that are being continually being improved upon (e.g. solar,
hydropower, onshore wind), many new technologies are being developed to expand the
possibilities of energy capture. Of these new methods (e.g. wave energy, water current (tidal)
energy and offshore wind energy), the offshore wind energy development (herein referred to as
offshore wind) market has made large strides globally and is presently proving its viability. As
the technology improves and the ability to deploy offshore wind turbines in deeper waters and
transmit that energy to shore becomes increasingly feasible, developers and agencies will likely
continue to look to expand on the offshore wind market.
At present, the world leader in offshore wind development and power generation is Europe,
which owns 90% of the installed capacity (EWEA, 2013). China and Japan are a distant second
and third representing 9% and 1%, respectively. Within the European community, the UK has
the largest amount of installed offshore wind capacity at 59%, followed by Denmark, Belgium,
Germany, the Netherlands, Sweden, Finland, Ireland, Norway and Portugal. Further, Spain has
recently deployed its first offshore wind turbine in 2013 (Dailyfusion, 2013). Much of the
European offshore wind development is due to the European Council directive that sets a
mandatory target of 20% share of energy (per country) by renewable sources by 2020.
As of the end of 2012, there were a total of 1,662 installed and grid-connected offshore wind
turbines in Europe, with a potential to generate 4,995 MW of power. This is an increase of 31%
from 2011 (EWEA, 2013). In the coming years, capacity is expected to continue to increase
another 66% in Europe as the cumulative capacity approaches 8.3 GW by 2014.
In the U.S. there is more onshore wind capacity installed than any other country (DOE, 2012),
but there is currently only one active offshore wind installation (though there are thirteen projects
that are in various stages of permitting and development [Trabish, 2013]). In June, the
VolturnUS 1:8 scale floating offshore wind turbine, designed and built at the University of
Maine, was successfully connected to the US power grid (http://composites.umaine.edu).
States with other offshore wind projects in queue include Massachusetts (Cape Wind and Wind
Energy Center), Rhode Island (Block Island and Wind Energy Center), Maine (Hywind Maine
Pilot Project and Aqua Ventus), New Jersey (Atlantic City), Delaware (Mid-Atlantic Wind
Park), New York (Long Island), Virginia (Virginia Beach), Ohio (Icebreaker project on Lake
Erie), Texas (Gulf Offshore Wind and Rio Grande North and South Projects) and Oregon
(WindFloat Pacific Demonstration Project).
The potential benefit of offshore wind energy in the U.S. is clear: according to the DOE, twenty-
eight U.S. coastal and Great Lakes states use approximately 78% of the U.S.’s electricity
(www.usoffshorewind.org). Further, it is believed that around 1/3 of all U.S power demand can
be satisfied from offshore wind resources along just the East Coast, alone (Biron, 2013).
3.1. Introduction There are many commercial and recreational users of U.S. coastal waters; and, offshore wind
development will need to seamlessly co-exist with these stakeholders. Also, the coastal
environment can be harsh and for the offshore wind industry to succeed, offshore wind structures
and infrastructure (foundations and cables) must not only survive, but thrive, with minimal
maintenance requirements.
With offshore wind foundation and cable stability of primary importance to turbine lifetime
longevity, scour, the net erosion of sediments around these structures, has become a highly
studied phenomenon. Laboratory and field experiments have resulted in empirical and semi-
empirical equations to determine the breadth and depth of scour holes that will form near a flow
obstruction, evaluations of which have provided some valuable guidance for future offshore
wind development (Cooper and Beiboer, 2002; den Boon et al., 2004; OSPAR, 2004; CEFAS,
2004; CEFAS, 2008a; CEFAS, 2008b; OSPAR, 2008; Stahlmann and Schlurmann, 2010; Yang
et al., 2010; DNV, 2011; among others).
These investigations have comprised mostly of scaled laboratory investigations, though, and
applications of the empirical equations to prototype scales have yielded mixed results. However,
the offshore wind development that has taken place in recent years has led to the accumulation of
field data (through established post-construction monitoring programs) and to the creation of
guidance frameworks that incorporate the latest advances in site planning and understanding.
These recommendations can be used by developers to streamline the permitting, planning, design
and development processes. Due to the infancy of offshore wind in the U.S., little is known about
either (a) the short- and long-term, and near- and far-field, environmental effects associated with
offshore wind installation, operation, and maintenance activities, or, conversely, (b) the effect the
environment has on the sub-aqueous wind turbine components during typical and extreme
(storm) conditions. Therefore, it is critical for the project planners to review all available
guidance literature, to be able to reasonably predict the ocean’s impacts on the wind turbine
structures, and to predict the wind turbine structure’s influence on the coastal processes, in order
to support efficient siting, planning and permitting efforts.
The motion of the ocean is largely driven by waves and tides. These processes will generate
forces throughout the water column that are important for foundation and cable structural
stability design. Moreover the near bottom currents generated by these coastal processes impart
stresses on the ocean floor sediments and are directly responsible for the erosion, transport and
deposition of these sediments.
Shallow water offshore wind technologies often require foundations systems that are anchored
through the ocean floor and/or resting on the seafloor surface, with transmission cables laid upon
or buried in the sediments. Transitional and deepwater offshore wind technologies often do not
involve traditional supports in or on the seafloor; however they tend to require additional
anchorage, stabilizing structures, or may comprise floating platforms that are anchored to the
26
seabed (Figure 2). The presence of each of these structures, their anchorages and their cables will
change local flow patterns in different manners, which may, in turn, alter local seabed dynamics.
Scour occurs where the sediment is eroded from an area of the seabed in response to the forcing
by waves and/or currents (Whitehouse, 1998). It has the potential to negatively affect offshore
wind foundation stability, cable installations, and critical attachment points (e.g. j-tubes),
requiring costly maintenance or repair. If a large number of offshore wind structures are placed
offshore of a coastline, near-shore wave energy and current circulation patterns may also be
adversely affected. Consequently, the way sediments are mobilized and distributed along the
coast and the manner in which wave action is imparted on a shoreline may be altered. Potential
consequences include generation of erosional “hot spots” and/or depositional “shoaling”
locations, disruption of natural littoral sediment transport, water quality alterations and
ecological consequences.
Figure 2. Examples of offshore wind turbine foundation types (www.offshorewind.net).
As mentioned, sediment in the coastal zone is primarily mobilized and transported by the action
of waves and currents. The overall sediment transport of a region is quantified in a sediment
budget, which describes the rate and amount of sediment being transported into and out of that
region. In a broad sense, regions may comprise large sections of a coastline and continental shelf
or a small embayment between two promontory headlands. A sediment budget, once defined, can
determine the type(s) and magnitude(s) of regional transport taking place and whether a coastal
region is ultimately eroding or accreting.
27
The sediment transport type is commonly categorized into cross-shore (perpendicular to shore)
and alongshore (parallel to shore) components and is directly dependent upon the direction of the
incident waves and currents. The sediment transport magnitude, therefore, is the amount of
sediment that is eroded, mobilized and re-deposited, and is dependent upon the strength and
duration of the incident waves and currents.
The key to understanding sediment transport is the identification, description, and quantification
of the dominant physical processes involved in moving sediments, and understanding how the
processes interact at a site. Although other seabed properties and characteristics may affect
sediment transport, an understanding of these fundamental physical processes is critical to the
overall quantification of transport. In the sections below the properties of site hydrodynamics
(waves and currents) and sediment characteristics that have the greatest influence on sediment
transport are established and their relative importance described.
3.2. Role of Hydrodynamic Processes Waves are typically generated by strong winds imparting stresses on the water surface over long
distances. The larger the wind stress (i.e. wind speed), and the longer that wind stress is applied
to the wave surface, the larger the wave height generated. The wave height that is generated is
typically limited by the duration for which the wind blows, the local water depth (depth-limited),
or the distance over which the wave has to grow (fetch-limited). Swell waves are generated by
winds that blow for long durations and over large fetches (distances). Sea waves are generated by
winds that blow for short durations or over short-fetches. Swell waves are generally created by a
distant storm and travel for long-distances. Sea waves are created by local storms and travel for
relatively shorter distances. Swell wave frequency and direction spectra tend to be narrow-
peaked and approach from a more focused direction; sea wave frequency and direction spectra
tend to be broad-peaked and may approach from multiple directions. Finally, waves are also
generated by other physical processes such as storm pressure gradients, storm surges, tide
fluctuations, seismic events (e.g. tsunamis) and transiting marine vessels.
Needless to say, wave motion is very complex. The combination of all wave-generating
processes mentioned above may create waves that are well-focused in a particular direction or
waves that are broadly spread in a wide direction (or multiple directions), which has a direct
effect on the resultant wave energy and sediment transport at a particular location. As deepwater
waves approach the coast, they are transformed by additional processes including refraction (as
they pass over changing bottom contours), diffraction (as they propagate around solid objects
such as headlands and breakwaters), shoaling (wave height increases as the depth decreases), and
energy dissipation (due to sea surface white-capping, seabed bottom friction, seabed vegetation,
and, ultimately, by wave breaking).
Combine these effects with the impacts of near-shore circulation (upwelling, lunar tide and/or
Coriolis effect), local flow and water level effects such as wind setup, storm surge and river
discharge, and long-term effects such as sea level rise, and the equation becomes increasingly
more complex. Though the prediction of hydrodynamic conditions may be difficult, the long-
term measurement and subsequent analyses of these data allows statistical assessment of many of
these processes at a particular site. It is the duty of the prudent engineer to determine the most
28
important hydrodynamic processes to evaluate at a site to describe the local wave regime with
the highest confidence level possible.
Figure 3. Udden-Wentworth grain size scale for sediments.
29
3.3. Physical Properties of Sediment For most systems, knowledge of particle size distribution and bulk density are instrumental to the
understanding of local sediment transport processes. Particle size (or grain size) distribution is
the most widely used property in engineering and environmental studies for the characterization
of the sediment bed. Sediment particle sizes are classed from very fine clays with a particle
diameter of 0.24 μm (.000 000 24 m) to boulders larger than 0.25 m in diameter. In between
these extremes are particle sizes that make up the sediment beds of common aquatic systems:
sands and silts. Figure 3 describes the typical ranges of particle (or grain) size associated with
each classification, along with a corresponding phi (Φ) classification that is also used in many
engineering and environmental classifications. The classification system shown here is
commonly referred to as the Udden–Wentworth classification system.
Sediments are generally classified as cohesive (fine grain-sized silts and clays) or non-cohesive
(coarse grain-sized sands and gravels). Cohesive sediments are sediments in which inter-particle
forces are significant, creating an attraction or cohesion between particles. Though there may be
some variation in definition, cohesive sediments are generally defined as those with particle sizes
less than 200 µm in diameter. The smaller ranges of cohesive particles (<62µm) are silts and
clays, and the larger sizes (62 to 200 µm) are fine sands. Non-cohesive sediments are those in
which inter-particle forces are insignificant, and are generally defined as those with particle
diameters larger than 200 µm. These size ranges typically include, at the smallest diameters, fine
to medium sands, and at the largest diameters, gravels.
Most often, natural sediments consist of a mixture of these sediment grain sizes, and samples are
often qualitatively described based on the relative proportions of each sediment type. For
example, a mixture of a small amount of sand with clay can be called sandy clay; and a smaller
amount of silt with larger amount of sand might be called a silty sand. As intuition suggests,
sediment mixtures will, therefore, have varying amounts of cohesiveness. This is an important
distinction because the cohesiveness of a sediment sample directly affects its susceptibility to
erosion, transport and deposition.
Bulk density is another basic property of a sediment bed that is useful for classifying sediments
and quantifying transport properties. The bulk density, ρb, of a sediment bed describes the overall
degree of packing or consolidation of the sediments, and can be described as the dry bulk density
or saturated (wet) bulk density. It is defined as the ratio of the mass of dry sediment to the total
sample volume (dry bulk density) or the total sample mass (sediment plus water) to the total
sample volume (wet bulk density).
The approximate particle density of the quartz and clay minerals that make up the majority of
sediment particles in the natural world is about 2.65 g/cm3, though some variation exists where
other sediment types and organic materials are encountered. The sediment bed, as mentioned, is
often composed of a mixture of different types and sizes of sediment particles packed into a
porous bed. In non-cohesive sediments, bulk density may not vary much with depth, as non-
cohesive particles typically do not readily compact over time. In cohesive sediments, though,
bulk density generally increases with depth into the sediment because the deeper sediments are
typically more consolidated; containing less porous space between individual particles. Cohesive
30
sediment beds will consolidate over time due to the weight of overlying sediment, expelling pore
water to the surface and bringing the particles closer together in the absence of the expelled
water. This causes an increase in the bulk density at increasing depth into the sediments. As the
bulk density increases due to consolidation, the potential for scour or erosion of the sediment
generally decreases as a larger shear stress is required to mobilize the compacted sediments
(Jepsen and Lick, 1997; Mehta and McAnally, 1998).
3.4. Sediment Erosion Erosion, water column transport, and deposition are the sediment transport processes that occur
in aquatic systems (illustrated in Figure 4). Erosion is defined as the flux (i.e. movement) of
particles from the sediment bed into the overlying water column. Deposition is the settlement of
particles out of the water column. Sediments that are in motion travel as bedload (bouncing along
the bottom), suspended load (up in the water column supported by turbulent mixing), or as wash
load (very fine particles dispersed throughout the water column that travel with the mean flow).
Figure 4. Simplified diagram of sediment transport processes.
Resting sediment bed particles are in equilibrium between the drag forces from fluid shear, the
lift forces from flow over the particles, gravitational pull on the particles, particle to particle
contact forces and cohesive inter-particle forces. At a certain velocity, though, the combined drag
and lift forces on the uppermost particles of the sediment bed are great enough to dislodge them
from their equilibrium positions. This velocity is termed the critical velocity and is proportional
to the critical shear stress for erosion, τce (measured in units of force per unit area (N/m2), which
is the shear stress at which sediment motion is initiated. This motion initially tends to occur only
at a few isolated spots, but, as the shear stress increases with increasing flow velocity, the
31
movement of particles becomes more widespread, causing a net erosive flux from the sediment
bed.
The flow velocity near the bottom can be modeled as a logarithmic profile in its simplest sense:
At the sediment bed, the velocity is zero (following standard pipe flow theory). Velocity
increases logarithmically above the bed until a distance is reached where the bottom friction no
longer affects the flow. The layer between the bed and this elevation, where the shear stresses
(e.g. friction) are the highest, is called the boundary layer (Figure 5).
Figure 5. Simplified diagram of sediment transport processes.
In riverine systems a unidirectional current is generally responsible for the shear stresses
imparted on the sediment bed. In coastal regions and estuaries, though, a combination of
oscillatory waves, currents, and fluctuating tides are responsible. Determination of the processes
responsible for imparting shear stresses on the sediment bed is important in characterizing a
sediment transport regime. Shear stress can be measured directly and/or indirectly in the
laboratory or in the field. It has been studied in detail for currents and waves, and can be defined
and quantified mathematically if given sufficient information about the hydrodynamics of the
system.
Finally, in addition to ambient forcing mechanisms (background wave and current forces),
erosion may also be induced due to enhancements of each of these phenomena. For example, as
flows encounter waterway constrictions and obstacles, the local flow speed typically increases in
proximity to these impediments. The localized flow increases may cause corresponding increases
in near-bottom shear stresses, which, if larger than the critical shear stress of the bed sediments,
may induce additional erosion.
32
If significant enough, the localized erosion (also known as scour) may cause structural
degradation or failure of the obstacle. In particular, erosion around wind turbine foundations,
cables, and other infrastructure has the potential to eventually undermine foundations and lead to
unexpected maintenance and system failure if left unchecked.
Scour around underwater structures subsequently becomes a sediment source as the sediment
becomes suspended in the water column, is transported away from its resting location, and is re-
deposited elsewhere. Net erosion occurs if, over time, the amount of sediment removed from the
bed in an area exceeds the amount that is episodically deposited.
3.5. Sediment Transport Once sediment has been mobilized, the subsequent transport mechanisms are divided into two
general modes: bedload transport and suspended load (which also includes wash load) transport.
Coarser particles and aggregates (or particles subjected to shear stresses similar to the critical
shear stresses) move along the bed by rolling and/or saltation (i.e., bouncing) in a thin layer as
bedload, whereas finer particles (or particles subjected to large enough shear stresses) are
suspended into the water column and move as suspended load. The mode of transport for a given
particle is largely affected by the sediment properties and flow regime of the region.
Bedload can account for a significant amount of sediment transport in systems comprised of
coarse-grained sediments (sands and larger), where the flow is high enough to cause rolling
motion but not strong enough to lift particles off of the sediment bed. Although bedload transport
may be dominant in coarse-grained rivers and near-shore coastal regions, it may or may not be of
significance in fine-grained (fine sands and smaller) regions such as estuaries, lakes or deeper
coastal waters. In fine-grained sediment systems, both individual particles and aggregates will
erode and move along the bed as bedload or suspended load depending upon the flow regime.
Individual particles may flocculate (cohere) during transport, increasing their likelihood and/or
rate of deposition; or the larger aggregates may break up into smaller aggregates or individual
particles during transport, making them more likely to travel as suspended load for a longer
period of time (Figure 6).
Figure 6. Cohesive aggregates eroded from the bed may disaggregate downstream.
33
Sediment particles transported as suspended load tend to move at, or very close to, the mean
velocity of the fluid. In a steady-state situation, upward turbulent transport of a sediment particle
by the fluid is balanced by the gravitational particle settling. This balance keeps the sediments
suspended in the water column. As long as the flow remains sufficiently turbulent, sediments
will be transported as suspended load. As current velocity decreases, suspended sediment
concentrations generally increase near the bed as heavier particles begin to settle out of
suspension. Vertical profiles of suspended sediment concentrations can be calculated based on
particle size, a reference suspended sediment concentration near the sediment bed, and the
ambient fluid velocity (Rouse, 1938; van Rijn, 1993). They can also be measured and estimated
with acoustic and/or optical instrumentation. This can be useful in determining overall sediment
flux of a region.
Within the water column, two processes generally dominate the movement and net transport of
particles: advection and turbulent diffusion. Advection is the transport of particles caused by the
motion or velocity of the fluid (i.e. mean current velocity). Turbulent diffusion is the dispersal of
particles in the water column due to random turbulent motion within the fluid. An accurate
characterization of these processes in any aquatic system will yield a good quantitative
description.
3.5.1. Non-Cohesive Sediment Transport
The mobilization of non-cohesive sediment (e.g. sands) is presently fairly well understood and
accepted by the scientific community. It is a function of the individual sediment particle grain
size diameter and the lift and drag forces (from the overlying current) being imparted on the
particle. In the simplest sense, when the lift and drag forces exceed the particle weight, the
particle will mobilize and begin to roll along the bottom (bedload transport). As higher flow rates
are imparted upon the sediment bed, the increasingly turbulent flows may cause sediment to be
suspended and transported in the water column (suspended load transport) before falling out of
suspension in another location (deposition).
The shear stress force imparted by the flow on the particle at the moment the particle mobilizes
is the critical bed shear stress. Shields (1936) developed an approach to relate the threshold of
non-cohesive sediment motion to the bed shear stress. It was a ratio of the force exerted by the
bed shear stress acting to move a grain on the bed, to the submerged weight of the grain
counteracting this (Soulsby, 1997): Eqn. 1
dg s
crcr
)(
where,
Θcr = the Shields parameter (dimensionless)
τcr = the threshold bed shear stress (Pa or N/m2)
g = the acceleration due to gravity (9.81 m/s)
ρs = grain density (kg/m3)
ρ = water density (kg/m3)
d = grain diameter (m).
34
Shields found an empirical relationship between the dimensionless Shields parameter and the
Reynolds number: Eqn. 2
v
du*Re
where,
v = kinematic viscosity
*u = the shear velocity defined as
Eqn. 3 2/1
*
bu
and τb is the bed shear stress applied by the flow.
The oft-referenced Shields diagram (Figure 7) is a plot of the Shields parameter versus the
Reynolds number. The curve was hand-drawn by Shields using all available sediment data at the
time, which only considered sediment under the influence of current shear forces. Using the
diagram, a user can infer whether a particular sized non-cohesive particle will be susceptible to
mobilization and transport given a specific flow regime.
Figure 7. Shields curve for the initiation of motion for steady flow (from ASCE 1975).
This theory has been further researched by Soulsby (1997) who, instead, plotted the Shields
parameter against the non-dimensional grain size, D (Figure 8). This procedure simplified the
solution process by avoiding the iterative process required by Shields’ method (i.e. where u*
appears in both axes). He also expanded the plotted dataset to include shear stresses imparted on
35
particles by waves, currents and the combined action of waves and currents. Soulsby and
Whitehouse (1997) then created an analytical solution (e.g. an equation) that mimicked Shields’
(hand-drawn) curve. When plotted with their additional data points, though, their equation (i.e.
the Shields curve) over-predicted the shear stress parameter values for very fine grain sizes (see
lower D* values in Figure 8).
Subsequently, Soulsby and Whitehouse altered the equation to account for the deviation at fine
grain sizes to yield an improved formula for predicting the threshold bed shear stress. As seen
below this curve is equivalent to their original formulation (i.e. Shields’ curve) at D* greater than
10, but more closely follows the data points for fine grained sediments.
Figure 8. Threshold of motion of sediments beneath waves and/or currents. The fitted curves of both Shields (1936) and Soulsby and Whitehouse (1997) are shown.
3.5.2 Cohesive Sediment Transport
Though studies on non-cohesive sediments have shown a strong correlation between particle size
and sediment transport rates, this observation does not hold for cohesive sediments, where
particle size alone cannot be used to predict transport rates (van Rijn, 1993; Roberts, et al., 1998;
Mehta and McAnally, 1998; Mehta, Hayter, Parker, Krone, and Teeter, 1989). The transport of
cohesive sediments (very fine sands, silts and clays) is also dependent upon properties and
characteristics such as the sediment bulk density, the organic vs. inorganic content in the
sediment, electrostatic and electrochemical forces, ambient water quality (e.g. salinity) and the
bioturbation activity. Cohesive particles tend to erode in aggregates made up of individual
sediment grains and/or flocculate during transport, both of which make the accurate prediction of
their erosion, transport and deposition characteristics difficult to predict. Further, the
cohesiveness and susceptibility of mobilization of a cohesive sediment source is site-specific,
36
requiring local knowledge and insight to evaluate the transport likelihood. Without the benefit of
empirical relations that can be applied universally to cohesive sediments, scientists and engineers
have resorted to collection of site-specific erosion data when needed.
There are presently several methods of directly measuring surface sediment erosion rates
(McNeil et al., 1996; Briaud et al., 2001; Roberts and Jepsen, 2001; Jepsen et al., 2002; Roberts
et al., 2003; Black, 2010; Rutgers, 2011). Some characteristics of these methods include in-situ
and ex-situ erosion rate measurements; measurement of bedload and suspended load fractions;
and erosion rate with depth below the sediment surface. Each has its own advantages depending
upon the overall project objective.
One method that has been employed frequently for the purpose of evaluating sediment erosion
rates below the surface is the Sediment Erosion at Depth Flume (SEDFlume; McNeil et al.,
1996). One distinct advantage of the SEDFlume is that it provides a means to directly measure
and quantify the erosion rates at distinct depths within a sediment core and for various applied
shear stresses. Using the measured data, engineers can then evaluate the likelihood of a site’s
sediments to erode given typical and extreme flow conditions. An example of how these data can
be utilized is briefly described here.
Following the methods of Roberts et al. (1998), the erosion rates of all natural sediments can be
approximated by: Eqn. 4
mnAE
where,
E = erosion rate (cm/s)
τ = bed shear stress (Pa)
ρ = sediment wet bulk density (g/cm3)
A, n and m = experimentally determined constants that depend on the sediment characteristics.
The constant, n, is always positive, implying that as the shear stress force increases, the
corresponding erosion rate of the sediments will increase. The constant, m, is always negative,
indicating that as bulk density increases (e.g. as a result of higher compaction at greater depths
into the sediment core), the erosion rate of the sediments decreases (i.e. sediments become more
difficult to erode). This holds true as long as all other sediment properties remain the same in the
core, which is an assumption that may not always be true in natural settings.
For large negative values of the constant, m, the sediments comprise of a large amount of
cohesiveness. The converse is also true: for small negative values of the constant, m, the
cohesive forces are weaker (i.e. the sediment is more non-cohesive). At a value of m = 0, the
sediments are non-cohesive.
Through direct measurement, the bulk density and erosion rate for a given shear stress can be
estimated. A least squares regression solution will yield the constant parameters A and n for that
sediment sample. Then, using a user-defined critical erosion rate threshold (e.g. 10-4
cm/s), the
corresponding critical shear stress can be determined. Conversely, when presented with a
37
specific shear stress or a time series of shear stresses, the erosion rate(s) of that sediment sample
can also be estimated.
3.5.3 Sediment Deposition
Deposition is the process by which sediment particles settle out of suspension onto the sediment
bed, causing an accretion of particles. As suspended and bedload sediments are transported, they
will encounter areas of lower fluid velocity. When sufficiently low velocity fluid is encountered,
turbulent eddies may be insufficient to keep the particles suspended or in motion as bedload and
the particles will settle to the sediment bed and motion will be halted.
The shear stress at which settlement begins is termed the critical shear stress for suspension, τcs,
and is also measured in units of force per unit area (N/m2). In a non-moving fluid, where no
shear stress is present, deposition rate is dependent solely on the settling speed of the sediment
particles and the sediment concentration in the overlying water. In flowing water, however,
deposition is affected by the fluid turbulence and near-bottom shear stresses which makes its
estimation difficult.
To quantitatively determine deposition rates at a specific location, one method incorporates a
probability of deposition, P, into the formulation to account for effects of the near-bottom shear
stresses: Eqn. 5
CwPD s
where D is the deposition to the sediment bed (g/cm2/s), C is the sediment:water concentration
(mg/L), ws is the particle settling speed described by Cheng (1997) as: Eqn. 6
5.12* 52.125 d
d
vws
and d* is the dimensionless particle diameter described by Blake et al. (2007): Eqn. 7
3/1
2
* 1
g
dd s
.
In this formulation, which estimates the settling speed of the particles based on the sediment
diameter, d is the median particle diameter (cm), ρs is the density of the particles (generally
assumed to be 2.65 g/cm3), g is the acceleration due to gravity (980 cm/s) and υs is the kinematic
fluid viscosity (cm2/s).
The probability of deposition, P, would be unity (i.e. 1) in the case of zero flow, and would
decrease as the shear stress increases. This formulation accounts for the decreased chance for
deposition as the shear stress increases. For sediment particles, Krone (1962) found that the
probability of deposition varied approximately as:
38
Eqn. 8
cs
cs
cs
P
1
0
.
When the shear stress is larger than the critical shear stress for suspension, τcs, no deposition will
take place. When the shear stress near the sediment bed is lower than τcs, particles will begin to
deposit onto the sediment bed proportionally (Blake et. al, 2007). As the shear stress decreases,
the probability of a particle settling onto the sediment bed and remaining there increases. At a
shear stress of zero, the probability of deposition is one (i.e. 100%).
Significant deposition can occur in deeper and/or less energetic coastal or lake environments,
where fluid velocity is very low or negligible. Furthermore, as fine-grained particles interact in
the water column, they may flocculate to form larger clumps that will settle out of suspension
faster than individual particles. This process is dependent on sediment type, suspended sediment
concentration, fluid velocity, ambient shear stresses, and water chemistry.
As the shear stress fluctuates in a natural system, the sediment bed may be subjected to episodic
erosion, deposition, and re-suspension. Net deposition occurs if, over time, the amount of
sediment deposited on the bed exceeds the amount that is episodically eroded.
3.6 Coastal Sediment Transport Processes
As previously mentioned, coastal sediment transport can be separated into alongshore and cross-
shore (or offshore) components. These processes can be differentiated one step further to reflect
the nature of the coastal zone in the near-shore (inside the wave breaking zone) and offshore
(outside the wave breaking zone). Although offshore wind farms are, by design, typically
installed outside the breaker zone, their potential effects extend to the near-shore and shoreline
and, therefore, warrant a discussion of the full coastal environment.
3.6.1 Near-Shore Coastal Zone
As waves approach shallow water regions cross-shore (shore-perpendicular) and alongshore
(shore-parallel) mass flux (and, therefore sediment transport) is induced by the breaking waves
and currents. The turbulence generated by the breaking waves suspends sediments into the water
column that are then transported by the ambient near-shore currents (which may be driven by
tidal currents, wave mass flux or other forcing mechanisms such as river discharge).
The cross-shore component of the mass flux will cause flow and sediment transport in an
onshore/offshore direction. The alongshore component of the mass flux will generate near-shore
currents parallel to shore, forcing sediment transport along the shoreline (Figure 9). Standard
coastal engineering methods can be implemented for determining the rate of cross-shore and
alongshore (a.k.a. littoral) sediment transport, which will be dependent on the size of the incident
waves (i.e. energy dissipation), direction of the waves, local shoreline configuration, near-shore
bathymetry and the sediment characteristics, among other properties.
39
Figure 9. Illustration of longshore current generation and resultant sediment transport.
Figure 10 illustrates the interaction of processes that cause one form of cross-shore flow, a rip
current, which is a localized region of offshore flow at a shoreline. The onshore flow (termed
"return flow" in the diagram) of water generated by the mass flux of breaking waves causes a
build-up (wave setup) of water mass on the shoreline. The imbalanced wave setup results in a
flow of water (rip current) that heads offshore in distinct locations. These rip currents are capable
of transporting mobilized sediment from the near-shore to outside the breaker zone; the stronger
the waves (i.e. larger mass flux), the more intense this process.
Typically, coastal regions experience a net onshore transport of sediment during low energy
periods (e.g. milder, summer conditions) and then a net offshore movement of sediment during
the higher intensity wave periods (e.g. strong, winter storms), which cause strong rip currents
and undertow currents. This is often reflected in larger beach widths observed during low energy
summer months; and, conversely, smaller beach widths observed during higher energy winter
months.
If there is a sufficient understanding of the typical wave and current conditions for a particular
region an overall net transport rate can be approximated. The balance of the sediment transport
due to cross-shore currents, alongshore currents, rip currents, and sediment exchange at the
boundaries characterizes the near-shore zone sediment transport for the location of interest.
40
Figure 10. Diagram of the primary, cross-shore, wave driven current in the near shore.
3.6.2. Offshore Coastal Zone
The offshore coastal zone, outside of the breaking wave zone, is typically where most offshore
wind farms will be deployed. The three largest coastal regions of the United States are the
Pacific, Gulf, and Atlantic Coasts (though the Great Lakes are also under consideration for
offshore wind development). When defining the regional boundaries for a planned project site,
the offshore boundary of these regions is generally considered to be the continental shelf.
Particularly on the Pacific Coast where the continental shelf is relatively close to the shore, the
boundary of the area of study is easy to define as the continental shelf. The Gulf of Mexico
offshore boundary is more difficult to define, though, as the shelf break may be much further
offshore of a planned development site than practical for consideration of transport processes. In
these cases, the offshore boundary should be located some logical distance where transport
processes can be assumed to have negligible impact on the offshore wind array. These distances
must be considered on a site-specific basis and may require iteration. The Atlantic Coast is a mix
of the two, where the shelf is both near and far from the shoreline, depending upon location.
Sound engineering judgment must be used on a site-specific basis for selection of these
boundaries.
41
As discussed, near-shore coastal surface waves are frequently a combination of wind-generated
sea waves and storm-generated swell waves that drive the near-shore circulation. In the deeper
offshore waters, the surface waves continue to have an impact on sediment transport; however,
this impact diminishes as the water depth increases. Typically, an offshore wind farm should be
located far outside the depth of closure, the depth at which sediment motion as a result of near-
shore wave and current forcing is negligible. Doing so will minimize mitigation measures
associated with wave-forcing mobilized sediment. But it will not eliminate all concerns: cable
installation costs increase with distance from shore and power transmission lines that cross the
near-shore region will still likely require engineering mitigation in order to minimize risk of
damage to the lines.
Circulation, as a result of surface waves and tidal circulation, is another forcing mechanism that
requires consideration offshore of the breaking wave zone. Tidal circulation results from the
movement of water due to the propagation of the tide through a region. The typical dominant
period of astronomical tides is 12.42 hours, yet the tidal range and variation are highly variable,
temporally and spatially. Coriolis forces (due to Earth’s rotation), storm surges and other local
meteorological weather patterns (e.g. variable pressure systems) may cause additional
fluctuations to the ambient circulation.
Furthermore, in the vicinity of large estuaries and rivers the offshore region may experience local
high-flow currents particularly during ebbing tides and large rainfall runoff events. Seasonal
cycles of temperature changes, fresh water input, and large scale winds can also substantially
affect the resulting offshore current circulation and wave climate. The interaction of the tides and
other processes result in a current structure that oftentimes dominates sediment transport in the
offshore.
Unlike near-shore sediment transport, which varies significantly on both short- and long-term
time scales, the sediment transport patterns on the continental shelf are generally in long-term
equilibrium with the prevailing wave and current climate. While there are still temporal and
spatial variations at all time scales, the sediment tends to organize into regular patterns (e.g.
bedforms) that are indicative of the long-term dominant configuration. These patterns can
develop and fade on short time scales in the near-shore (e.g. seasonal sandbars). Typically the
dominant spring tidal currents will develop a net movement of sediment along the shelf that is
periodically interrupted by storms that generate large waves and currents and direct large
sediment loads to the shelf. The disruption in the pattern by these events is quickly incorporated
back into the long-term dominant pattern. Figure 11 illustrates the range of processes that
interact in the offshore coastal zone.
42
Figure 11. Illustration of sediment transport processes interacting in the offshore coastal
zone.
43
4. EROSION AND SCOUR FUNDAMENTALS
4.1. Foundation Obstructions
Oftentimes the primary risk to an offshore wind structure (and its peripheral components) placed
in a natural flow regime will be the risk to structural stability created by localized erosion or
scour. Scour is the net removal of sediment from the vicinity of the structure foundation (or
components) that increases susceptibility of structure (or component) failure. Scour may have an
impact on the geotechnical capacity of a foundation and thereby on the structural response that
governs the ultimate and fatigue load effects in structural components (DNV, 2011). Most
foundations are generally oriented perpendicular (vertical) to the seabed and are known as
‘vertical obstructions’ as they obstruct the flow regime vertically through the water column. As a
result, flow streamlines much transition horizontally around the object.
Scour is a consequence of flow obstruction caused by the structure; the very presence of the
structure alters the ambient wave and current flow streamlines (i.e. flow must alter direction
around the structure), creates wake vortices, and leads to an increase both in the speed of the
flow in the vicinity of the structure, and in the turbulent intensity of the flow (Whitehouse, 1998;
DNV 2011). The increase in flow speed near the structure is a result of the conservation of mass
(Continuity Theory): the flow is being constricted and must accelerate around the obstruction.
The increase in near-bed turbulent intensity is a consequence of the generation of flow vortices
around the structure. Both velocity and turbulent intensity amplifications create intensifications
in near-bottom shear stresses, ultimately increasing the likelihood of sediment erosion and
mobilization.
In the case of solidly supported offshore wind structures (e.g. monopoles, jacket structures and/or
gravity based foundations) scour may erode sediment that is providing vertical and lateral
support; the loss of which may lead to an increase in bending stresses unless remedial action is
taken (Watson, 1979, from Whitehouse, 1998). Scouring typically produces seabed depressions
adjacent to the structures, reduces the effective depth of pile penetration, and may expose
suspended risers, anchors, or other components to hydrodynamic loading that exceeds design
limits. For suction caissons or mat foundations, scour may reduce the weight of sediment acting
against the overturning moment of the structure and lead to overturning instability. Scour effects
dissipate with increasing distance from the structure as the flow becomes less affected.
Scour is commonly classified as live-bed or clearwater scour. Live-bed scour occurs when the
threshold necessary to mobilize sediment is exceeded everywhere on the bed (e.g. a high-flow
river regime causing universal sediment movement); sediment transport proceeds from upstream
to downstream and through scour depressions, if any exist. It is overall seabed movement;
regional, large-scale erosion, deposition and bedform movement through a region; and ambient
morphology. Oftentimes, there is no net erosion or deposition as a result.
44
Clearwater scour occurs when the upstream, ambient flow is insufficient to mobilize sediment;
yet, flow speed amplifications resulting from the structural obstruction are sufficient to cause
nearby sediment mobilization and erosion. The presence of the structure causes a sufficient
increase in flow velocity and shear stress to facilitate erosion. Clearwater scour may occur
adjacent to bridge piers at a river crossing during typical river flow regimes.
Clearwater scour results from the flow disturbances directly generated by the flow obstruction.
The seabed boundary layer flow approaching a vertical cylinder (e.g. pile), for example, creates a
pressure gradient on the upstream face of the cylinder between the low pressure in the near-bed
flow and the high pressure in the flow above. This drives a flow downward at the face of the pile.
A primary vortex is formed at the upstream face of the pile during this stage as the downward
flow impinges with the seabed. The vortex then wraps around the cylinder creating the secondary
horseshoe vortex, and trails off downstream (Whitehouse, 1998). Secondary vortices are
periodically shed from either side of the cylinder as the flow diverges around the obstruction
(Figure 12).
Figure 12. Hydrodynamics around a slender pile with scour.
Clearwater scour may be further characterized as:
Local scour – where steep-sided scour pits form adjacent to individual piles or slender
obstructions, or
45
Global (or dishpan) scour – where shallow, wide depressions form under and around
individual or groups of structures and obstructions such as jacket foundations (
Figure 13).
In addition to near-bed turbulence from the vortices, the cross-sectional area of flow around the
cylinder is constricted, resulting in a corresponding increase in flow speed around and near to the
structure (adhering to the definition of Continuity Theory). Therefore, sediment particles are
likely mobilized by the increase in turbulence and then transported downstream by the increase
in flow speeds. From a sediment transport perspective, the primary and secondary horseshoe
vortices (and increased velocities) are the major mechanisms leading to the scouring of sediment
from around the base of a cylinder (Whitehouse, 1998). Similar vortices are generated and shed
by alternatively-shaped obstructions (e.g. square piles, rectangular, diamond, oval). The
obstruction shape and orientation to the incident flow have a direct effect on the manner in which
vortices, and resultant scour patterns, are generated.
Furthermore, the vortex generation and resultant scour processes described above also pertain to
multiple-piled structures (such as that illustrated in
Figure 13) and large-scale offshore flow obstructions (e.g. wind farm deployments). Each of
these types of obstructions acts to disrupt the ambient flow to varying degrees, possibly resulting
in varying magnitudes of clearwater and global (dishpan) scour near the structure. The
magnitude of scour will depend upon several factors such as ambient current speed, incident
current direction, amount of flow that is obstructed, the total size of the obstructing structure and
the proximity of multiple piles formulating the total obstructed area.
Figure 13. Example of local (clearwater) and global scour around a structure (from Whitehouse, 1998; reproduced from Angus and Moore, 1982).
46
4.2. Cable and Pipe Obstructions
Similarly to large structural obstructions, pipelines laid along the seabed, pipes extending short
distances horizontally from subaqueous offshore wind structures (generally parallel to the
seabed) and the corresponding cables protruding from them that are laid along the seabed also
act as flow obstructions, and may experience adverse effects of scouring. If extender pipes
(sometimes termed J-pipes) and cables are not sufficiently buried in the seabed, or become
exposed due to clearwater scour, flow separation will occur in flows passing over the pipeline as
flow passes above and below the obstruction. This results in an area of re-circulating flow being
produced in lee of the obstruction.
Eddies may be shed from the pipeline and cause fluctuating shear stress (and erosion) in lee of
the pipeline. Since extender pipes extend only short distances from the infrastructure to which
they are connected, they, and the cable they support, are within the zone susceptible to scour pit
formation caused by the larger vertical obstruction (e.g. monopile), and are more likely to be
exposed to the scour processes described here. If J-tubes are exposed to ambient flow, the cable
within may be allowed to freely move with the local currents. Overtime, the cable can wear
against the end of the J-tube (and the sediment that can accumulate there), abrading the cable,
potentially leading to failure.
Pressure gradients between the upstream and downstream sides of pipes and/or cables (as a result
of flow divergence and separation) resting on the seafloor may induce a seepage flow in the sand
bed underneath the pipe/cable, called the ‘onset of scour’ (Sumer and Fredsoe, 2002).
Eventually, the seepage flow may cause a mixture of sediment and water to “break-through” the
space beneath the pipe/cable in a process termed ‘piping’. Piping proceeds rapidly to ‘tunnel
erosion’ as the flow begins to diverge beneath the pipe/cable and scour additional sediments
(Figure 14). As a result of the Continuity Theory, very high flow velocities beneath the
pipe/cable will exist initially due to the small cross-sectional area of space through which the
flow can proceed. As more sediment erodes, the void beneath the pipe/cable will continue to
grow (both in depth and width). If the tunnel erosion beneath the pipe/cable is sustained,
eventually a free span will be formed which may leave the pipe/cable susceptible to altered
Potential sources of sediment to the coastal zone (rivers, inlets)
Current magnitudes and directions (tide ranges, frequency)
Wave environment (typical and extreme wave heights, periods and directions)
Other local information (storm surge, sea level rise, tsunami potential)
Environmental concerns (contaminated sediment/water, listed and endangered species)
When developed, the CSM will indicate any data gaps and the relative importance of those data
can be assessed. The full CSM is typically a concise written document with basic maps denoting
information gained during the initial data collection and analysis phase. This document can
provide information not only for determining impacts, but assisting the offshore wind design,
construction, and operation teams.
5.1.4. Wind Farm Characteristics
To supplement the development of the CSM, preliminary offshore wind farm design parameters
should be defined and included to assist with the CSM evaluation. This is in preparation for
assessment of likely environmental effects of the offshore wind farm. Most wind turbine
installation schemes are similar to those used in the oil and gas industry: solid supports
(monopoles, tripods and/or jacket structures) driven into, or resting on, the seafloor; gravity
based foundations, which rely on their own weight to remain fixed to the seafloor); moored,
floating systems; or, some combination of systems. The type of placement and fixing technique
is dependent on the physical characteristics of the deployment locations (i.e. water depth,
namely), balanced by the cost of design and installation.
Solid support systems are deployed in water depths ranging between 10 and 30 m that can
typically be found within a few miles of the US coastline. Gravity based foundations are
typically deployed in shallow coastal regions where the equipment to drive piles for solid point
connections have difficulty accessing the site. Both of these technologies are deployed in
relatively shallow waters where sediment mobility and scour can be significant concerns. Also,
the relatively shallow water depths where these foundation technologies are generally deployed
regularly correspond to locations close to the US coastline (within a few miles), often within
visual range of the shoreline, and potentially in high vessel traffic areas. As such, important
impediments to their use in US waters are visual pollution and navigation safety (DTI, 2005a;
DTI, 2005b).
Moving further from the coast and into transitional and deeper waters requires investigation of
alternative deployment schemes, but can also cause installation costs to become prohibitive.
Although less explored, moored systems seem to alleviate some of these concerns. Large floating
structures of this type were first developed by the offshore oil and gas industry and the
technology can be directly applied to wind turbines. Placement of offshore wind turbines further
offshore has a twofold benefit: the wind is typically stronger and more consistent during daylight
81
peak energy times, and the negative aesthetic effect of turbine visibility to coastal residents is
mitigated. The offshore wind industry is currently investigating moored systems in water depths
larger than 100 m.
5.1.4.1. Solid Support Systems
Solid support systems are based on hard connections, often a hollow pile (or multiple piles in the
case of tripods or jacket structures) or a concrete-filled pile(s) extending from the turbine to the
seabed. Solid support systems may also include gravity-based turbines, which are anchored to
the seafloor with a large weight. Solid support systems require a sufficient connection with the
ground (or anchor to the ground), otherwise the turbine and its support system will move
irreversibly. The foundation must transfer the forces from the structure to the surrounding soil. It
is critical that the foundation sustain all loads that may be applied, particularly during extreme
environmental conditions (e.g. wind and waves during storms) to reduce maintenance and/or
replacement costs associated with structural failures. As structures are subjected to repeated
loading (whether by ambient hydrodynamic forces or augmented forces due to scour and loss of
support) structural stiffness degradation becomes an important consideration. The structural
design must consider the cumulative lifetime stresses to which the support members may be
subjected.
The types and approaches for hard connections widely vary and are chosen based on sediment
bed characteristics including soil conditions such as sand density and depth to the clay stratum,
as well as the strength of the underlying clay. Solid support systems can be as simple as
concreting a monopile into holes drilled into the bedrock (Figure 26). This approach was
successfully deployed in Blyth, Northumberland, UK.
Monopiles work well in shallow waters with hard bottoms, but are not suitable for loose, mobile
sand banks, glacial till and soft clay. These types of sites require different types of foundation
such as suction caisson multi-foundation structures (jacket structures) and suction caisson
monopod structures (Figure 26). Jacket structures prove more economical for use when it is
favorable to displace structural load only on surface sediments. These foundations contain
perimeter ‘skirts’ embedded into the sea floor so that the effect of scour is mitigated. Overturning
loads applied by the wind and waves in Jacket structures are resisted predominantly by equal and
opposite vertical loads at foundation level. In this design, the foundations are likely embedded in
sand where the response of the foundation to vertical loads is critical. For monopiles, the
overturning load is applied directly to the single large foundation in the form of shear and
moment; as opposed to traditional axial compression and tension loading methods. Additional
variations of wind turbine foundations are likely under development; only the basic solid support
foundation types are discussed here to provide representative examples.
82
Figure 26. Solid support systems used for offshore wind applications: (a) monopile structure; (b) suction caisson multi-foundation structure; (c) a suction caisson monopod
structure.
Gravity based foundations (GBFs) offer an alternative to pile and jacket structures. They are
often most advantageous in shallow waters where pile-driving is not a feasible option. GBFs can
be floated or towed to location and anchored in place, without the need to anchor into the seabed
substructure. At the simplest level, they often consist of a larger, heavily weighted base resting
on, or partially beneath the surface sediments. Figure 27 shows a sample GBF design.
83
Figure 27. Example gravity based foundation (http://www.eon-uk.com).
5.1.4.2. Floating Systems
Floating wind turbines fall into two main categories: 1) tension leg mooring, and 2) Catenary
mooring. Tension leg mooring systems have vertical tethers under tension providing large
restoring moments in pitch and roll. The vertical tethers are taut and run relatively straight to the
sea bed (Figure 28).The stabilizing tension in the tethers results from a heavy sea bed anchor on
the seabed side, and a large buoyant force on the topside (floating platform). Catenary moorings,
on the other hand, get their restoring force through the weight of the chain and steel shackles,
and not necessarily line tension. The tether lines are not as taut as those seen in tension mooring
systems, and often arc to the seabed. Catenary systems provide station-keeping yet provide little
stiffness at low tensions. Higher tensions and ballasted catenaries can be used to increase the
stiffness and stability of these types of systems.
Floating moored structures, primarily as a result of development in the oil and gas industry, have
matured to allow installation in depths well over 1000 meters, permitting the development of
offshore wind turbines in vast stretches of ocean. Installation and maintenance costs for these
systems must strongly be considered for a viable project. The key design parameters of a floating
wind turbine platform, whether tension or catenary moored, is the selection of the optimal
combination of floater shape and size, ballast weight, and mooring attributes (angle and tension
of mooring lines or chains). The goal is to keep the floater responses within acceptable bounds of
pitch, roll, and heave, yet minimize construction and installation costs.
84
Figure 28. Top panel shows a general catenary moored platform where the tethers are arced. Bottom panel shows an exaggerated general tension leg moored platform where
the tethers are drawn taut.
Figure 29. Pictorial representation of a moored floating wind turbine. The main components of the support system are the nacelle, platform, ballast and mooring system.
The support system of a floating wind turbine can be described by its main components: the
nacelle, platform, ballast, and the mooring system (Figure 29). The platform geometry is defined
by the barge radius and draft. It gives rise to the buoyancy force required for tension mooring
systems, and provides the necessary connection and foundation for the turbine itself. The bottom
Turbine Nacelle,
85
side of the platform is often weighted with a concrete ballasted steel cylinder to achieve static
floatation stability in pitch and roll. The mooring system is defined by an anchor and the tether
lines or chains that connect the platform to the anchor. Water depth, line tension and the angle
between the free surface and the anchor line segment are key parameters in mooring design.
Mooring systems often consist of grouped tethers that are evenly spaced around the platform to
enhance stability.
5.2. Impact Assessment
Once developed, the CSM provides the launch pad for a thorough site analysis and
environmental impact assessment of installing an offshore wind farm (or other obstructing
structures) at a particular location. The following sections focus on specific considerations and
methods for evaluating the seabed stability, and other environmental impacts, as a result of
offshore wind installations, but can easily be applied to other alternative energy installations.
They are organized to provide the reader with a representative sample of impact considerations
such that a preliminary assessment of changes to the hydrodynamics and sediment transport
patterns caused by the offshore wind farm can be completed. However, as previously mentioned,
these may not be all-inclusive; and a thorough site-specific determination of important
considerations should be made. Once general hydrodynamic and sediment dynamics changes are
understood, the impact analyses can be iteratively focused on specific areas of concern for both
the design (e.g. near-field, fine-scale scour development) and the environment (e.g. far-field
sediment transport pattern changes and/or ecological changes). These final impacts can then be
described and mitigated for in the wind farm planning and development phases.
5.2.1. Site Impacts
Once the CSM has been constructed to the satisfaction of site planners and managers, a site
analysis can be completed which evaluates the baseline condition (existing scenario) with project
development alternatives. The result should be a comparison of before and after hypothetical
scenarios that project the potential positive and negative impacts to a site based on the alteration
of site characteristics and physical processes. From the perspective of this guidance document,
the two objectives of the site analysis should be:
Quantitatively evaluate the local coastal forcing mechanisms and response to installation
of an offshore wind farm
Quantitatively evaluate the local morphological reaction to installation of an offshore
wind farm.
These objectives and comparison can be addressed through analytical methods, physical
modeling and/or numerical modeling. In the example in Appendix A, a numerical model
example has been created to show utility of this option in completing an initial site analysis of
before and after construction scenarios. It is a simple model created as a basic example of a
potential site analysis. Reality may require more detailed investigations.
86
5.2.2. Environmental Impacts
Some common environmental concerns for offshore wind farms include noise production,
94. Yang, R-Y, H-H Chen, H-H Hwung, W-P Jiang and N-T Wu. (2010). Experimental Study on
the Loading and Scour of the Jacket Type Offshore Wind Turbine Structure. Coastal
Engineering Proceedings, No 32.
108
109
APPENDIX A: EXAMPLE COASTAL HYDRODYNAMICS AND SEDIMENT DYNAMICS MODEL
An example model from the Monterey Bay and Santa Cruz, CA, coastline is used here to
illustrate the utility of combining the SWAN wave propagation and EFDC circulation/transport
models to predict near-shore sediment stability in the presence of an offshore wind array. A
coarse-grid regional wave model of Monterey Bay was established (domain shown in Figure 32)
within which a finer resolution grid model was nested to assess near-shore impacts in proximity
to Santa Cruz. The overall modeling approach described herein has the following limitations:
It is a simplification of a turbulent, chaotic, near-shore process.
Coriolis forces, salinity and temperature gradients are not included at the offshore
boundaries. In other words, large scale (e.g. CSM scale, regional scale, scales larger than
the littoral cell) ocean circulation is not incorporated into the near-shore region.
Measurements of currents were only available at near-shore locations for model
validation.
Even though the above limitations are considered when assessing the results, this methodology
produces reasonable estimates of transport when forced with the dominant near-shore processes
in the region (i.e. wind, waves and tides).
A.1. Models
The SWAN wave propagation model is a non-stationary (non-steady state) third generation wave
model, based on the discrete spectral action balance equation and is fully spectral (over the total
range of wave frequencies). Wave propagation is based on linear wave theory, including the
effect of wave generated currents. The processes of wind generation, dissipation, and nonlinear
wave-wave interactions are represented explicitly with state-of-the-science, third-generation
formulations. Model boundary conditions can be explicitly specified by the user or may be
obtained from nested, larger-domain modeling efforts (either a larger SWAN domain, or other,
global models such as WaveWatch III). SWAN allows for numerous output quantities including
two dimensional (frequency and direction) spectra, significant wave height, mean wave period,
mean wave direction and bottom orbital velocities (due to wave oscillations). The SWAN model
has been successfully validated and verified in laboratory and complex field cases worldwide.
The hydrodynamic model, EFDC (Environmental Fluid Dynamics Code), is an US EPA
approved, state-of-the-art, three dimensional hydrodynamic model developed at the Virginia
Institute of Marine Science by John Hamrick (1992) to simulate hydrodynamics and water
quality in rivers, lakes, estuaries, and coastal regions. The EPA describes the model as “one of
110
the most widely used and technically defensible hydrodynamic models in the world.” This model
has the following capabilities and features:
The model is 3-dimensional, which allows for the simulation of variations in current
structure in the vertical as well as horizontal.
It allows input of near-shore wave radiation stresses and wave energy dissipation for
simulation of surf zone circulation and transport.
The model allows incorporation of complex bathymetry.
The model allows input of time varying flows, winds, water levels, and discharges.
To accurately model the transport of particles in the coastal environment, it is critical to describe
both the transport and the bottom shear stress. EFDC handles advective transport using the
modeled water column velocities. These velocities are computed from tidal forces, wave forces,
and wind.
EFDC uses the Smagorinsky (1963) method to calculate the horizontal diffusivity. The
magnitude of the diffusivity in the model is proportional to the horizontal current shear. The
dissipation of wave energy can be calculated in the SWAN wave model and used as an input to
EFDC. The wave dissipation then acts as another source of turbulence and can be added to the
KH determined from the currents in the Smagorinsky model.
EFDC implements the Mellor and Yamada (1982) second moment turbulence closure model in
the vertical orientation. The model, as implemented in EFDC, has been improved and further
validated by Galperin et al. (1988). Once the vertical diffusivity has been calculated through the
Mellor and Yamada and Galperin model, the wave dissipation from the SWAN model is added
in as a source of turbulence. The wave and current generated bottom shear stresses can then be
calculated using the Christoffersen and Jonsson (1985) formulation.
A.2. Setup and Validation
The first phase of any modeling analysis is to verify that the model is functioning correctly and
also reasonably simulating the natural processes occurring at the site. To ensure that the model
closely simulated currents in the project area, measured wave and current data were compared
with modeled values. The SWAN results were validated with nearby NOAA NDBC buoy wave
data (Figure 33). Output wave conditions from the SWAN model were incorporated into the
EFDC model and measured tide and winds were applied to the EFDC domain. The EFDC results
were validated with near-shore measured current velocities (Figure 34).
111
Figure 35 illustrates the peak wave heights simulated within the near-shore Santa Cruz EFDC
model.
Figure 36 shows an expanded view of the modeled wave heights with superposed velocity
vectors from the study area. These results indicate along shore velocities propagating to the east
and are consistent with previously conducted drifter observations and ADCP measurements
collected during a field measurement period (Chang et al., 2010). The combined wave and
current shear stresses and velocities derived from the coupled SWAN/EFDC model provide the
fundamental physical parameters for assessing both environmental and design impacts of an
offshore wind farm deployed in this region.
Figure 32.Monterey Bay model domain and bathymetry. NOAA NDBC buoys used for model validation are shown in green.
112
Figure 33.Model (line) representing the wave height (Hs), peak wave period (Tp) and mean wave direction (MWD) obtained from the Monterey Bay SWAN model. Measured
data (dots) were obtained from the NOAA NDBC buoy 46236 in Monterey Bay.
Figure 34. Model (line) representing the current magnitude obtained from the nearshore Santa Cruz EFDC model. Measured data (dots) were obtained from a Teledyne RDI ADCP
deployed during the field study.
113
Figure 35. Peak wave heights modeled using SWAN in the Santa Cruz, CA region. Area of interest highlighted by red outline.
See Expanded View Below
114
Figure 36.Peak wave heights and velocity vectors in the model domain. Expanded view of the above region.
A.3. Simulating Offshore Wind Devices
For the example modeling effort, individual support structures for a 200-turbine wind farm were
simulated in the SWAN model, centered on 40 meter water depths. For this model each device
was simply considered a monopole structure with a 10 meter diameter. Devices were spaced at
50 meters, center to center (i.e. 30 meter separation from device to device). It is acknowledged
that these water depths and dimensions may be in disagreement with some design standards.
This geometry and location was selected for example purposes, to show functionality of the
model. This methodology, device size, spacing, layout and location can be customized for any
planned installation configuration.
The SWAN model allows for multiple methods of obstructing wave energy; for this effort,
offshore wind turbines were simulated as discrete obstructions to the propagating wave energy.
A coefficient of reflection and transmission were specified, which dictated the percentage of
115
wave energy that was allowed to be reflected and propagated past the obstructions. To simulate
an extreme scenario, wave energy was not reflected and was completely blocked from
transmission at each obstacle. In essence, all wave energy was absorbed by the obstructions
creating an obvious gradient in wave energy in lee. Specifying wave energy blockage in this
manner is a relatively simple specification using existing SWAN functionality and capability.
The locations of the wind turbines defined in the model are illustrated in Figure 37. The wave
heights predicted by the model both before and after offshore wind installation are shown in
Figure 38. The most notable effect of the inclusion of a large 200-turbine array is that wave
heights are substantially reduced in lee of the structures. This is due to the simulated absorption
and of wave energy by the simulated foundations. The change in wave patterns as a result of the
obstructions can be incorporated into the sediment transport assessment to examine both near-
and far-field effects due to the presence of an offshore wind farm.
Figure 37. Monterey Bay model domain. 200 offshore wind turbine array.
116
Figure 38. Modeled wave heights before (top) and after (bottom) the inclusion of a wind turbine array.
117
SWAN modeling runs were initiated using a combination of typically occurring, and extreme
event, wave conditions as the boundary conditions. Modeling was completed in stationary mode
during the technique development because of the smaller sized modeling domain (non-stationary
processes were considered negligible at the present time, but can be easily incorporated at a later
date). Model outputs included wave heights, wave periods and wave directions for the entire
modeling domain. In addition, the near-bottom orbital velocities (due to shallow water, non-
linear wave oscillations) were exported and included in near-bottom shear stress computations.
Wave-driven currents were combined with current flow in EFDC and near-bed shear stresses
were computed as a result of the combined action of waves and currents. The computed shear
stresses were then used with site-specific sediment size information to estimate sediment
mobility and susceptibility to erosion.
A.4. Evaluating Sediment Risk
As discussed, the movement of sediment in the coastal zone is dynamic and varies spatially and
temporally. The scenario presented here describes a basic evaluation of the changes in sediment
movement that may occur due to the offshore wind farm. When planning an offshore wind farm,
more detailed evaluation of the site-specific impacts of the offshore structures might be
necessary. These may include a fine-scale, detailed analysis of the near- and far-field scour and
sediment transport potential as well as the prediction of any disruption to the natural sediment
transport patterns.
To characterize seabed sediment mobility it is important to characterize the sediment and its
spatial distribution in the system, as well as the near-bed shear stresses generated by the local
waves and currents. Here, knowledge of the spatial distribution of sediment grain size and
combined wave/current generated shear stresses at the site was used to establish an initial
understanding of sediment mobility.
Sediment particle size distribution data were interpolated to the same model computational grid
domain used in the hydrodynamic analysis in a GIS. Sediment phi sizes, where available, were
converted directly to grain diameters and were assumed to be the median grain sizes:
Eqn. 21
d2log
where d is the sediment diameter, in millimeters.
Near-bottom shear stresses due to the wave activity were computed following the method of
Christoffersen and Jonsson (1985), which accounts for the ambient current velocities, wave-
induced orbital velocities and seabed roughness. Figure 39 shows sample results of shear stress
calculations both before (baseline scenario) and after installation of wind turbine arrays. The
sediment roughness used in the model is the individual median grain size of the gridded particle
size distributions, and is based on the measurements reported by the USGS (Reid et al., 2006).
118
The bed shear stresses were computed and transferred to GIS for rapid evaluation and
visualization of regional bed shear stresses. If the critical shear stress of the sediment is known
with some degree of certainty, then the spatial location of the sediments can be classified
according to the sediment erodibility likelihood. This is a function of the critical shear stress of
the sediments and the expected (modeled) bed shear stress under the given wave boundary
conditions.
The shear stress ranges were ordered into a 10-point magnitude risk scale. The estimated critical
shear stress of the sediments at the site was established as the mid-point of the 10-point scale.
Spatial areas with predicted shear stresses lower than this value had a low risk of mobilization.
Areas with shear stresses higher than this value ere more susceptible to erosion (i.e. are at a
higher risk of mobilization). A classification of 10 implied the sediments were highly susceptible
to mobilization; a classification of 1 indicated the sediments were not very susceptible to
mobilization, and, may, in fact, be more susceptible to deposition. The final results are illustrated
on a sediment stability risk map for easy visualization of the risk or potential for sediment
transport (Figure 40).
The baseline sediment stability risk map (absence of turbines) is extremely valuable, in that
offshore wind developers can identify areas of natural high and low probability of sediment
movement prior to installation, and therefore avoid high risk areas for both foundation
deployment and cable routes. In the presence of offshore obstructions, the structures generally
reduce the likelihood of sediment transport in the area in lee of the array by reducing wave
activity behind the structures. As a result of the wave energy pattern changes modeled here, there
may be an alteration of circulation in the region. The results suggest an array installed at this
location may induce sediment deposition behind the structures.
Deposition could result in the potential for habitat alteration (e.g. sea grass burial), both in
proximity to the array or further downstream near the shoreline. The far-field change in risk
along the coast, however, does not show any widespread alteration. It is important to note that
this report illustrates the expected bed shear stresses and associated sediment risk of sediment
mobilization resulting from one applied wave and current scenario and may not be representative
of all situations.
A more detailed quantitative analysis is required to fully evaluate the range of expected shear
stresses due to typical and extreme wave and current conditions. At the scale of this model,
sediment mobility caused by scour around foundations is not assessed. Foundation scour is a
critical design impact to consider and will be discussed more in the next section. Conceptually,
foundation scour pits will mobilize more sediment in the vicinity of the devices, making more
sediment available for deposition in the lee of the array.
Figure 41 illustrates the components of the general sediment transport assessment methodology
described in the preceding sections.
119
Figure 39. Modeled seabed shear stresses before (top) and after (bottom) the inclusion of
a wind turbine array.
120
Figure 40. Risk of sediment transport before (top) and after (bottom) the installation of a
wind turbine array.
121
Figure 41.Flowchart of risk assessment methodology.